Materials Transactions, Vol. 52, No. 6 (2011) pp.

1288 to 1293
#2011 The Japan Institute of Metals EXPRESS REGULAR ARTICLE

A Novel Process for Producing Ferromolybdenum Powder Master Alloy

without Generating Secondary Pollutants
through a Two-Step Hydrogen Reduction Process
Byung-Su Kim* , Sang-Bae Kim, Hoo-In Lee and Young-Yoon Choi
Minerals & Materials Processing Division, Korea Institute of Geoscience & Mineral Resources, Daejeon, Korea

A new process for producing ferromolybdenum powder master alloy without generating secondary pollutants by a two-step hydrogen
reduction process involving MoO3 and mill scale has been developed. In this process, mill scale which is an iron byproduct generated from the
steel rolling process is used as an iron source. Experimentally, ferromolybdenum powder master alloy was produced by a two-step hydrogen
reduction process: The first hydrogen reduction stage was carried at a horizontal furnace for 60 min at 848 K under a hydrogen gas at a partial
pressure of 101.3 kPa, and the second hydrogen reduction stage was done at the same furnace for 40 min at 1123 K under the same hydrogen
partial pressure. In the process, the reduction reactions of MoO3 to MoO2 and mill scale to Fe proceed simultaneously in the first reduction step,
and then the reduction reaction of MoO2 to Mo does in the second reduction step. The reaction rate of the second step in the two-step hydrogen
reduction process of MoO3 -Fe2 O3 mixture was also measured under isothermal condition in hydrogen atmosphere using TGA equipment. As an
example, at 1173 K under a hydrogen partial pressure of 101.3 kPa, almost 100% of MoO2 was reduced to Mo in 5 min. The nucleation and
growth model were found to be applicable to the reaction rate. The reaction was first order with respect to hydrogen partial pressure and had an
activation energy of 83.8 kJ/mol (20.1 kcal/mol). [doi:10.2320/matertrans.M2011037]

(Received January 28, 2011; Accepted March 4, 2011; Published May 25, 2011)
Keywords: ferromolybdenum powder master alloy, molybdenum oxide, hydrogen reduction, nucleation and growth model

1. Introduction
Ferromolybdenum is an important iron molybdenum alloy
with a molybdenum content of 60–70% because it is used as
the main source for molybdenum alloying of high-strength-
low-alloy (HSLA) steels in the steel industry.1) At the present
time, ferromolybdenum alloy is produced by either a thermite
process or a carbon reduction process.2–4) However, in the
processes large amounts of slag and dust which cause
environmental problems are inevitably generated. Specially,
the thermite process uses high cost aluminum and silicon as a
reducing agent of MoO3 .
Therefore, it would be highly desirable to develop an
alternative process to manufacture ferromolybdenum alloy.
A technology was suggested to manufacture ferromolybde-
num powder master alloy from the mixture of MoO3 with Fe
powder using the two-step hydrogen reduction process which
is used to product pure molybdenum metal powder.5) In the
two-step hydrogen reduction process, a MoO3 -Fe mixture
is first reduced by hydrogen at below 923 K, whereafter
the mixture obtained is manufactured to ferromolybdenum
powder master alloy by hydrogen reduction at around
1373 K. However, the process has never been commercial-
ized because hydrogen used as a reducing agent of MoO3 in
the process is expensive. Although that, the process has many
advantages because slag and dust are not generated in the
process and additional additives are not needed. Thus, there is
recently growing interest in manufacturing ferromolybdenum
alloy by the two-step hydrogen reduction process. However,
very little fundamental information about kinetic data on the
two-step hydrogen reduction process is available in the
literature.
The present research is concerned with developing a new
process for manufacturing ferromolybdenum powder master
*Corresponding author, E-mail: bskim@kigam.re.kr
alloy from a mixture of MoO3 with mill scale by the two-step
hydrogen reduction that is an environmental friendly process
compared with conventional processes. The proposed proc-
ess uses mill scale instead of Fe powder as an iron source and
reduces MoO3 and mill scale simultaneously. The process is
thus simple and economical compared with the previous
hydrogen reduction process. The mill scale is one of iron
byproducts with an iron content of over 70% generated from
the steel rolling process. In this paper, the rate data of the
hydrogen reduction of MoO3 -Fe2 O3 mixture was also
thermogravimetrically measured under isothermal condition
using TGA equipment. It is therefore expected that the
results obtained in this study will help to design the new
process for manufacturing ferromolybdenum powder master
alloy.
2. Experimental
Experiments were carried out in horizontal tube furnace to
determine the products from the MoO3 -mill scale reaction
system, and a typical thermogravimetic analysis (TGA)
apparatus similar to the one described in the previous paper
was used to measure the rate of the hydrogen reduction
reaction of MoO3 -Fe2 O3 mixture.6)
In the former experiments, a weighed amount of the
mixture of MoO3 with mill scale was placed in an alumina
boat (1.2 cm diameter and 8.0 cm length) which was pushed
into the central hot zone of a mullite reaction tube of 6 cm
diameter and 130 cm length that was maintained at a desired
temperature. The reaction system was made oxygen-free with
an argon gas flow inside the reactor before the reactant gas
was introduced at a controlled rate. The sample was heated
up to 848 K for the first hydrogen reduction, followed by the
second hydrogen reduction at 1173 K. During heating up to
1173 K after the first hydrogen reduction, the reactor was
purged with argon gas. The solid product cooled in argon
 A Novel Process for Producing Ferromolybdenum Powder Master Alloy without Generating Secondary Pollutants 1289

atmosphere after the second hydrogen reduction was weigh- 2500

Ο
ed, the product analyzed by X-ray diffraction (XRD) and 2000 Ο MoO2 MoO3 : 1,135 mg + Mill scale : 465 mg
scanning electron micrograph (SEM). ∇ Fe
1500
The TGA apparatus consisted of a recording microbalance FeO
Ο
1000

Intensity (A.U)
from one arm of which a shallow silica tray for sample was Ο Ο
suspended by a platinum chain into a reactor tube located 500 ∇ ΟΟ
Ο
∇
within vertical tubular furnace. The reactor tube was an 0
Ο
Inconel tube of 5.0 cm inner diameter and 65 cm length. 1200
MoO3 : 980 mg + Mill scale : 620 mg
The microbalance continuously recorded the mass changes
taking place during the reaction. During the experiments, 800
Ο
the balance was purged with argon gas to prevent the Ο Ο
400 ΟΟ
intrusion of reacting gas and hot gases. MoO3 -Fe2 O3 ∇ Ο
∇
mixtures weighing 690–710 mg placed in a shallow holder 0
were used in the TGA experiments. The mixture ratio was 20 40 60 80
fixed at a mixture ratio to product ferromolybdenum powder 2 Theta / degree
master alloy which contains 70 mass% Mo. The sample
Fig. 1 XRD patterns of the solid phases obtained from the first hydrogen
was maintained for 60 min at 848 K for the first hydrogen reduction step of MoO3 -mill scale mixture under a hydrogen partial
reduction, followed by the second hydrogen reduction at pressure of 101.3 kPa for 60 min at 848 K.
various temperatures. During heating after the first hydrogen
reduction, the reactor tube was purged with argon gas. The
reduction temperature and time in the first hydrogen
12000
reduction stage were respectively chosen through prelimi-
Mo
nary experiments.
10000
Hydrogen (99.9%) and argon gases (99.9%) were supplied
by Air Products Company in Korea. The MoO3 powder, 8000
manufactured by Aldrich Chemical Co., was 99.9 mass%
Intensity (A.U)

pure, and the Fe2 O3 powder (99.9 mass% pure iron) was 6000
obtained from Alfa Aesar Co. Both materials were of
74 mm size. Also, the mill scale that is one of byproducts 4000
generated from a steel rolling process of special steel Mo
company in Korea mainly contains 46.6 mass% FeO, 2000
Mo
49.9 mass% Fe2 O3 , 0.14 mass% Cu, 0.01 mass% P and
Fe2Mo Fe2Mo3
Al2 O3 , CaO, MgO, and SiO2 as minor amounts. Usually, it 0
was known that the control amount of impurities in
10 20 30 40 50 60 70 80
ferromolybdenum alloy is below 0.5 mass% for copper and
0.06 mass% for phosphorous because they have deleterious 2 Theta / degree
effects on the mechanical properties of the alloyed steel
Fig. 2 XRD pattern of the solid phase obtained from the second hydrogen
produced. It is thus considered that the mill scale can be reduction step under a hydrogen partial pressure of 101.3 kPa for 40 min at
used as an iron source for manufacturing ferromolybdenum 1123 K after the first hydrogen reduction step of MoO3 -mill scale mixture
alloy. The mill scale used for this study was 100 mm in size. under a hydrogen partial pressure of 101.3 kPa for 60 min at 848 K.

3. Results and Discussion

3.1 Reaction products
In order to verify the products from the MoO3 -mill scale
reaction system, several experiments were carried out in
horizontal tube furnace. For the two-step hydrogen reduction
of MoO3 -mill scale mixture, a hydrogen gas at a partial
pressure of 101.3 kPa was passed over the mixture held in an
alumina boat for 60 min at 848 K in the first hydrogen
reduction step and for 40 min at 1123 K in the second
hydrogen reduction step. The input amount was 980 mg of
MoO3 and 620 mg of mill scale for a low grade ferromo-
lybdenum powder master alloy and 1135 mg of MoO3 and
465 mg of mill scale for a high grade ferromolybdenum
powder master alloy.
(1) Products in the first step
Figure 1 shows the X-ray patterns of the solid phases
obtained from the first hydrogen reduction step of MoO3 -mill
scale mixture. It is seen that the solids contains only MoO2
and Fe, and partially reduced FeO. These results indicate that
the reduction reactions of MoO3 to MoO2 and mill scale to
Fe proceed simultaneously in the first reduction step.
(2) Products in the second step
The X-ray pattern of the solid obtained from the second
hydrogen reduction step of MoO3 -mill scale mixture re-
vealed only molybdenum and molybdenum-iron alloy, as
shown in Fig. 2. This result indicates that the reduction
reaction of MoO2 to Mo and the alloying phenomenon
between Mo and Fe powders reduced proceed simultaneously
in the second reduction step. Figure 3 shows the SEM picture
of a ferromolybdenum powder master alloy produced from a
MoO3 -mill scale mixture by the two-step hydrogen reduction
process. The SEM picture indicates that the product particles
are quite porous and the crystallites grow. Based on the
results, it was verified that ferromolybdenum powder master
alloy can be produced by the two-step hydrogen reduction
reaction between MoO3 and mill scale.
1290
Fig. 3 SEM picture of a ferromolybdenum powder master alloy produced
from a MoO3 -mill scale mixture by the two-step hydrogen reduction
process.
1200
1000
Temperature, T / K
800
600
400
200
0 2000 4000 6000 8000
Time, t / s
Fig. 4 A typical TGA curve for the two-step hydrogen reduction reaction
of MoO3 -Fe2 O3 mixture.
3.2 Kinetic analysis on the rate data of the second step
In the novel two-step hydrogen reduction process for
manufacturing ferromolybdenum powder master alloy from
MoO3 -mill scale mixture, the rate data of the second step
were measured using analytical grade Fe2 O3 powder instead
of mill scale as an iron source to minimize the effect of
impurities contained in the mill scale on the reaction rate.
Figure 4 shows a typical TGA curve for the two-step
hydrogen reduction of MoO3 -Fe2 O3 mixture. As shown in
Fig. 4, the rate change of weight loss was examined in the
first hydrogen reduction step. This result indicates that the
hydrogen reduction rate of Fe2 O3 is much faster than that of
MoO3 , as shown in Fig. 5. Figure 5 shows the reduction ratio
vs time relationships for the hydrogen reduction of pure
MoO3 to MoO2 and pure Fe2 O3 to Fe under a hydrogen
partial pressure of 101.3 kPa at 848 K. Here, the reduction
ratio at a particular time was determined by dividing the mass
change of the solid sample measured after reacting at the
time by the theoretical maximum total mass change. It was
thus analyzed in the first hydrogen reduction step that the
hydrogen reduction reaction of Fe2 O3 mainly occurs at the
first part and the hydrogen reduction reaction of MoO3 does
mainly at the second part. Based on the results, after the first
B.-S. Kim, S.-B. Kim, H.-I. Lee and Y.-Y. Choi
1.0
0.8
Reduction ratio ( - )
0.6
0.4
0.2 For Fe2O3 sample
For MoO3 sample
0.0
0 500 1000 1500 2000 2500 3000 3500
Time, t / s
Fig. 5 Reduction ratio vs time relationships for the hydrogen reduction of
pure MoO3 to MoO2 and pure Fe2 O3 to Fe under a hydrogen partial
pressure of 101.3 kPa at 848 K.
hydrogen reduction step, it was considered that the weight
loss detected at the second hydrogen reduction step is mainly
700 caused by the hydrogen reduction of MoO2 to Mo.
Rate measurements for the reduction of MoO2 in the
600 second step of the two-step hydrogen reduction reaction of
Weight (mg)
MoO3 -Fe2 O3 mixture, which is of the ultimate interest for
500 manufacturing ferromolybdenum powder master alloy, were
carried out at temperature ranges between 1023 and 1173 K
400 under hydrogen partial pressures between 10.1 kPa and
101.3 kPa. The experiments were continued until the solid
300 sample produced after the first hydrogen reduction reaction
showed no noticeable further mass change. The rate data
10000 12000
obtained were measured by minimizing the effects of
external mass transfer by using a sufficiently high flow rate
(8.3 mL/s) of hydrogen and small amount (690–710 mg) of
the MoO3 -Fe2 O3 mixture, which was chosen through
preliminary experiments. The reduction ratio at a particular
time was determined by dividing the mass change of the solid
sample measured after the first hydrogen reduction reaction
at the time by the theoretical maximum total mass change.
The theoretical maximum total mass change was calculated
by assuming that MoO2 in the solid sample produced after
the first hydrogen reduction reaction is totally reduced to Mo.
The assumption was considered to be reasonable because the
weight loss occurred in the first hydrogen reduction reaction
is very similar to theoretical maximum total loss caused by
the reduction reactions of MoO3 to MoO2 and Fe2 O3 to Fe,
and as shown in Fig. 6, the X-ray pattern of the solid obtained
from the first hydrogen reduction reaction reveals only MoO2
and Fe, and partially reduced FeO.
The effect of temperature on the reduction reaction of
MoO2 in the second step of the two-step hydrogen reduction
reaction of MoO3 -Fe2 O3 mixture was determined by varying
the reaction temperatures between 1023 and 1173 K, while
all other experimental variables such as sample mass and gas
flow rate were kept constant for the measurement. Figure 7
shows the results for the reaction of the second step in the
two-step hydrogen reduction of MoO3 -Fe2 O3 mixture under
a hydrogen partial pressure of 101.3 kPa. It is seen that almost
100% of MoO2 was reduced to Mo in 5 min at 1173 K under a
hydrogen partial pressure of 101.3 kPa. On the other hand, the
 A Novel Process for Producing Ferromolybdenum Powder Master Alloy without Generating Secondary Pollutants 1291

1.0
1600 Ο
Ο MoO2
∇ Fe

K
Reduction ratio, (-)
0.8

1173
FeO
Intensity (A.U)

1200

K
23
0.6

10
800
Ο 0.4

Ο
400 ∇ Ο
0.2 For MoO3 + Mill scale sample
Ο Ο
Ο ∇ For MoO3 sample
0.0
0 0 200 400 600 800
20 40 60 80 Time, t/s
2 Theta / degree
Fig. 8 Reduction ratio of MoO2 in the second step of the two-step
Fig. 6 XRD pattern of the solid phase obtained from the first hydrogen hydrogen reduction reaction of MoO3 -Fe2 O3 mixture and pure MoO3
reduction step of MoO3 -Fe2 O3 mixture under a hydrogen partial pressure powder under a hydrogen partial pressure of 101.3 kPa.
of 101.3 kPa for 60 min at 848 K.

1.0
1.0

0.8
2

2
Reduction ratio, XMnO

0.8
Reduction ratio, XMoO

0.6
0.6
PH (kPa)
2

0.4 10.1
0.4
Temp. ( K ) 30.4
1023 50.7
0.2 1073 0.2 70.9
1123
101.3
1173
0.0 0.0
0 200 400 600 800 0 1000 2000 3000 4000
Time, t / s Time, t / s
Fig. 7 Effect of reaction temperature on the reaction of the second step in Fig. 9 Effect of hydrogen partial pressure on the reaction of the second
the two-step hydrogen reduction of MoO3 -Fe2 O3 mixture under a step in the two-step hydrogen reduction of MoO3 -Fe2 O3 mixture under a
hydrogen partial pressure of 101.3 kPa. reaction temperature of 1123 K.

effect of the reduced iron on the reduction of MoO2 in the
second step was examined to be nearly negligible, as shown
in Fig. 8. Figure 8 shows the results for the reduction
reaction of MoO2 in the second step of the two-step hydrogen
reduction reaction of MoO3 -Fe2 O3 mixture and pure MoO3
powder under a hydrogen partial pressure of 101.3 kPa. These
results might be explained by the fact that as shown in Fig. 5,
the hydrogen reduction rate of Fe2 O3 powder is much faster
than that of MoO3 .
The effect of hydrogen partial pressure on the reaction was
examined by varying the hydrogen partial pressure between
10.1 kPa and 101.3 kPa in an argon atmosphere, while the
reaction temperature was kept constant at 1123 K. This was
done by varying the flow rates of hydrogen and argon gases
while maintaining the total flow rate at 8.3 ml/s. Figure 9
shows the effect of hydrogen partial pressure on the reaction.
These curves show that the reduction ratio increases with
increasing hydrogen partial pressure and almost 100% of
MoO2 is reduced to Mo in 50 min at 1123 K under a hydrogen
partial pressure of 10.1 kPa.
The interpretation of the rate data was performed using a
number of different rate equations including the spherical
shrinking-core model and power law, from which nucleation
and growth model proved to yield the best results. On the
other hand, several researchers reported that the nucleation
and growth model was found to fit the hydrogen reduction
reaction rate of metal oxides.5,6) The applicability of this rate
expression can be expected from the phenomenon that
hydrogen might be adsorbed on the MoO2 particles that form
nuclei like Hx MoO3 before reacting with MoO2 .7) In this rate
expression, the reduction ratio of MoO2 to Mo is related to
the reaction time by8)
½ lnð1  XMoO2 Þ1=m ¼ kapp  t ð1Þ
where XMoO2 is the reduction ratio of MoO2 , m is a constant,
kapp is the apparent rate constant (s1 ), and t is the reaction
time (s) which is given by
kapp ¼ b  k  f ðpH2 Þ ¼ b  k  pH2 n ð2Þ
Here, b is the stoichiometry constant (b ¼ 1=2 in this system,
according to the formulation of Sohn9)) for the hydrogen
reduction reaction of MoO2 to Mo, k is the intrinsic rate
constant (s1 kPan ), pH2 is the hydrogen partial pressure
(kPa) in the bulk gas, f designates the partial pressure
1292 B.-S. Kim, S.-B. Kim, H.-I. Lee and Y.-Y. Choi

2 0.010

K First order
2 3 0.008
11
ln [ - ln ( 1-X MoO ) ]
2

kapp / s-1
K 0.006
7 3
11 K
23
10
0.004
K
-2 73
10
0.002
Experimental data
Best fit data
-4 0.000
3 4 5 6 7 0 20 40 60 80 100
ln t / s pH / kPa
2

Fig. 10 Plots of ln½ lnð1  XMoO2 Þ versus ln t using the rate data of Fig. 7 Fig. 12 Dependence of the apparent reaction rate constant on the partial
according to eq. (1).
pressure of hydrogen from the results shown in Fig. 9.

2
-8.0

a a
0 kP kP E = 83.8 kJ/mol
4
ln [ - ln ( 1-X MoO ) ]

3 .
1. a 30
10 kP
2

.9 -8.5
ln K / S -1 ·kPa -1

70

-2 a
a kP
kP .1
.7 10
50 -9.0

-4
Experimental data
Best fit data -9.5
-6
3 4 5 6 7 8
8.5 9.0 9.5 10.0
ln t / s 4
10 / T / K
Fig. 11 Plots of ln½ lnð1  XMoO2 Þ versus ln t using the rate data of Fig. 9
according to eq. (1). Fig. 13 Arrhenius plot of the intrinsic rate constants obtained from the
results of Fig. 7.

dependence of the rate, and n is the reaction order for the

partial pressure of hydrogen gas. It is apparent from eq. (1) intercepts are plotted in Fig. 12 against the hydrogen partial
that a plot of ln½ lnð1  XMoO2 Þ against ln t should be linear pressure. As shown in Fig. 12, a straight line through the
with m as the slope and m ln kapp as the intercept for ln t ¼ 0. origin is obtained, which indicates a first-order reaction with
The validity of the nucleation and growth model rate respect to the hydrogen partial pressure in a mixture with
expression for the rate data of the second step obtained from argon. Thus, eq. (2) can be rewritten as
the two-step hydrogen reduction reaction of MoO3 -Fe2 O3
kapp ¼ b  k  f ðpH2 Þ ¼ ð1=2Þ  k  pH2 ðs1 Þ ð3Þ
mixture was verified by first plotting the reduction ratio-time
curves of Figs. 7 and 9 according to eq. (1), as shown in The values of the rate constant, k, were obtained from
Figs. 10 and 11. Examination of these figures reveals that the eq. (3) at various temperatures and plotted in Fig. 13.
rate data follow well eq. (1). In these Figs. 10 and 11, the Figure 13 is an Arrhenius plot of the intrinsic rate constants.
values of the slopes were calculated by regression analysis. The slope of the straight line placed through the experimental
The straight lines with high-correlation coefficient (r > points corresponds to an activation energy of 83.8 kJ/mol
0:993) were selected to represent the possible controlling (20.1 kcal/mol). The line through these data can be expressed
mechanism. The best-fit values of m at different hydrogen by the following equation:
partial pressures and temperatures totaling nine runs varied
k ¼ 1:35  exp½10;078:9=T ðs1 kPa1 Þ ð4Þ
between 1.48 and 1.70. The use of a single value of m for the
same material is consistent with the mechanistic justification Using eq. (4), the reaction rate of the second step in the two-
for the use of the nucleation and growth kinetics equation, step hydrogen reduction reaction of MoO3 -Fe2 O3 mixture is
and thus all of the calculated lines were plotted with an thus represented by the following equation:
average value of m to be 1.58.
½ lnð1  XMoO2 Þ1=1:58 ¼ kapp  t ð5Þ
In order to evaluate the hydrogen partial pressure depend-
ence of kapp , the values of kapp thus obtained from the with kapp ¼ 6:75 X 101  exp½10;078=T  pH2 (s1 )
 A Novel Process for Producing Ferromolybdenum Powder Master Alloy without Generating Secondary Pollutants
4. Conclusion
Experimental results showed that ferromolybdenum pow-
der master alloy could be produced from the mixture of
MoO3 with mill scale by a two-step hydrogen reduction
process. This process scheme is attractive because the
process can use mill scale instead of Fe powder as an iron
source and reduces MoO3 and mill scale simultaneously
without generating any solid waste. In the process, the
reduction reactions of MoO3 to MoO2 and mill scale to Fe
proceed simultaneously in the first reduction step, and then
the reduction reaction of MoO2 to Mo and the alloying
phenomenon between Mo and Fe powders reduced do
simultaneously in the second reduction step. The rate of the
second step in the two-step hydrogen reduction reaction of
MoO3 -Fe2 O3 mixture was found to be very fast. For example,
at 1173 K under a hydrogen partial pressure of 101.3 kPa,
almost 100% of MoO2 was reduced to Mo in 5 min. The
nucleation and growth model was applicable in describing
the kinetics of the reaction of the second step in the two-step
hydrogen reduction reaction of MoO3 -Fe2 O3 mixture over
the entire temperature range. The reaction was first order with
respect to hydrogen partial pressure and had an activation
energy of 83.8 kJ/mol (20.1 kcal/mol). The reaction rate of
the second step in the two-step hydrogen reduction reaction
of MoO3 -Fe2 O3 mixture is represented by the following
equation:
1293
½ lnð1  XMoO2 Þ1=1:58 ¼ kapp  t
with kapp ¼ 6:75 X 101  exp½10; 078=T  pH2 (s1 )
Acknowledgements
This research was supported by the General Research
Project of Korea Institute of Geosciences and Mineral
Resources (KIGAM) funded by the Ministry of Knowledge
Economy of Korea.
REFERENCES
1) R. Q. Barr: Kirk-Othmer Encyclopedia of Chemical Technology vol. 15,
3rd ed., (America: John Wiley & Sons, 1981) pp. 670–682.
2) Samsun Logix Corp., Seoul, Korea, private communication (February,
2008).
3) K. Jamshidi, H. Abdizadeh, S. A. Seyyed Ebrahimi and K. Hanai: Int. J.
Refractory Metals Hard Mater. 22 (2004) 243–245.
4) G. Li, D. Wang, X. Jin and G. Z. Chen: Electrochem. Commun. 9 (2007)
1951–1957.
5) M. M. Gasik and P. N. Ostrik: J. Thermal Anal. 40 (1993) 313–319.
6) B. S. Kim, E. Y. Kim, H. S. Jeon, H. I. Lee and J. C. Lee: Mater. Trans.
49 (2008) 2147–2152.
7) H. Sakagami, Y. Asano, T. Ohno, N. Takahashi, H. Itoh and T. Matsuda:
Appl. Cataly. A 297 (2006) 189–197.
8) H. Y. Sohn and S. Won: Metall. Trans. B 16B (1985) 831–839.
9) H. Y. Sohn: Metall. Trans. B 9B (1978) 89–96.